For this , you are asked to go through the following steps for coreference resolution: (1) Download GAP coreference repository from the link:
https://github.com/google-research-datasets/gap-coreference, (2) develop your program that resolves ambiguous pronouns that refer to names, and (3) assess the performance of your program against 4,000 pairs in gap-test.tsv evaluated by Precision, Recall, F-score, and Bias ratio with two task settings: snippet-context and page-context. Snippet-context indicates that the context (sentences) in the Wikipedia URL cannot be used, and pagecontext indicates that the context can be used. Bias ratio is calculated by taking the ratio of feminine (F) to masculine (M) F1 scores - F/M. Further information about the data can be found from the link above. Note that your program will be evaluated as gap_scorer.py.
As before, you should use techniques that can be implemented in Python and NLTK, scored as a relevance.
A Write a Python code for coreference resolution based on the repository.
Discuss your results, to explain how you addressed the goal and to suggest how you can improve the quality of the results further.
